Resolution enhancement and zoom by degradation estimates

Abstract

In this invention, the resolution is enhanced by first effectively decreasing the scan angle subtended between adjacent samples significantly below that of the Rayleigh limit to obtain an image blurred by the point spread function (or diffraction pattern) of the aperture. The next step is to process this blurred image to at least partially remove the blur. The unblurring process consists of convolving this finely sampled blurred image with a specially designed convolution mask. This mask effectively outputs an enhanced pixel at each step of the convolution. The mask simultaneously performs the equivalent of the following operations: (1) blur the image again; (2) subtract this reblurred image from the original image to form degradation estimates; and (3) add the estimates of the losses back into the original image.

What is claimed is: 
     
       1. In an imaging system comprising an imager which may view a scene through an aperture characterized by a point spread function and means responsive to said imager for generating successive frames of video data words characterized by a first sampling rate, a system for enhancing image resolution, comprising: means for converting said video data words into a synthesized video frame of video data words, said synthesized video frame characterized by a second higher sampling rate;   means for enhancing the resolution of the image represented by said synthesized video frame, comprising:   means for convolving said synthesized video data words with said point spread function to generate a video data word corresponding to a twice-blurred image;   means for subtracting said twice-blurred image video data word from a corresponding one of said synthesized video data words to obtain a video data word corresponding to a degradation estimate; and   means for adding said degradation estimate video data word and said corresponding one of said synthesized video data words to obtain an enhanced image video data word.   
     
     
       2. The system of claim 1 wherein said means for converting into a synthesized video frame comprises multiple image registration means for forming a mosaic of displaced ones of said successive video frames generated by said imager. 
     
     
       3. The system of claim 1 wherein said means for converting into a synthesized video frame comprises means for performing image interpolation in a single one of said successive video frames. 
     
     
       4. The system of claim 1 further comprising an array or detectors having a detector-to-detector sampling rate greater than the Nyquist rate. 
     
     
       5. The system of claim 1 wherein said enhancing means performs said convolving, subtracting and adding using a single convolution mask. 
     
     
       6. In an imaging system including an imager which views a scene through a system which has a limiting aperture characterized by a point spread function, a method for enhancing image resolution comprising: increasing the sampling density of a blurred image to produce a highly sampled blurred image;   convolving said highly sampled blurred image with said point spread function to generate a twice blurred image;   subtracting said twice blurred image from said highly sampled blurred image to obtain a degradation estimate; and   adding said degradation estimate to said highly sampled blurred image.